Severity first, then effort — week one always ships something reversible. Cumulative figures use the conservative end of every range.
The conservative (low) end of each shipped finding's range, summed through this week:
| Finding | Low end |
|---|---|
| LCP 12.2s on news | 0 |
| TBT 1,890ms on NBA | 0 |
| TBT 1,690ms on MLB | 0 |
Every row is an assumption-bounded range, not a forecast — roughly 10–30% of A/B tests produce a significant win. The full arithmetic for any row is on its finding page.
The conservative (low) end of each shipped finding's range, summed through this week:
| Finding | Low end |
|---|---|
| LCP 12.2s on news | 0 |
| TBT 1,890ms on NBA | 0 |
| TBT 1,690ms on MLB | 0 |
| LCP 9.3s on MLB | 0 |
| LCP 9.1s on NBA | 0 |
Every row is an assumption-bounded range, not a forecast — roughly 10–30% of A/B tests produce a significant win. The full arithmetic for any row is on its finding page.
The conservative (low) end of each shipped finding's range, summed through this week:
| Finding | Low end |
|---|---|
| LCP 12.2s on news | 0 |
| TBT 1,890ms on NBA | 0 |
| TBT 1,690ms on MLB | 0 |
| LCP 9.3s on MLB | 0 |
| LCP 9.1s on NBA | 0 |
| TBT 1,500ms on weather | 0 |
| TBT 1,140ms on news | 0 |
| TBT 1,130ms on article | 0 |
Every row is an assumption-bounded range, not a forecast — roughly 10–30% of A/B tests produce a significant win. The full arithmetic for any row is on its finding page.
The conservative (low) end of each shipped finding's range, summed through this week:
| Finding | Low end |
|---|---|
| LCP 12.2s on news | 0 |
| TBT 1,890ms on NBA | 0 |
| TBT 1,690ms on MLB | 0 |
| LCP 9.3s on MLB | 0 |
| LCP 9.1s on NBA | 0 |
| TBT 1,500ms on weather | 0 |
| TBT 1,140ms on news | 0 |
| TBT 1,130ms on article | 0 |
| LCP 5.4s on article | 0 |
| Missing trust signals | 0 |
Every row is an assumption-bounded range, not a forecast — roughly 10–30% of A/B tests produce a significant win. The full arithmetic for any row is on its finding page.